Bacillus Vallismortis Sp. Nov., a Close Relative of Bacillus Subtilis, Isolated from Soil in Death Valley, California
Total Page:16
File Type:pdf, Size:1020Kb
INTERNATIONALJOURNAL OF SYSTEMATICBACTERIOLOGY, Apr. 1996, p. 470-475 Vol. 46, No. 2 0020-7713/96/$04.00+0 Copyright 0 1996, International Union of Microbiological Societies Bacillus vallismortis sp. nov., a Close Relative of Bacillus subtilis, Isolated from Soil in Death Valley, California MICHAEL S. ROBERTS,'" L. K. NAKAMURA,' AND FREDERICK M. COHAN3 Center for Microbial Ecology, Michigan State University, East Lansing, Michigan 48824-1I01 '; Microbial Properties Research, National Center for Agricultural Utilization Research, Agricultural Research Service, US. Department of Agriculture, Peoria, Illinois 61 604'; and Department of Biology, Wesleyan University, Middletown, Connecticut 06459-01703 Five Bacillus strains isolated from Death Valley soil were shown to belong to a previously unidentified species, for which we propose the name Bacillus vallismortis. The type strain is strain DV1-F-3 (= NRRL B-14890). On the basis of previously published restriction digestion data, B. vallismortis is most closely related to Bacillus subtilis. At this time B. vallismortis can be distinguished from B. subtilis only by differences in whole-cell fatty acid compositions, DNA sequences, and levels of reassociation of genomic DNA. The species most closely related to Bacillus subtilis are easily cludes the prefix B-, which identifies organisms that were obtained directly from distinguished by differences in their DNA sequences. B. subti- individuals or were isolated at the National Center for Agricultural Utilization Research, or the prefix NRS-, which identifies strains obtained from the N. R. lis, Bacillus mojavensis, Bacillus atrophaeus, Bacillus amyloliq- Smith Bacillus collection.) uefaciens, and Bacillus lichenifomis can be distinguished by Selection for spontaneous rifampin-resistantmutants. We obtained rifampin- genomic hybridization data (11, 12, 16); these species form resistant (Rif; encoded by rppoB) mutants of each strain (which were used as distinct DNA sequence clusters for shared genes (15, 16), and donors in transformation experiments) by using previously described methods (16). as a result of sequence divergence, levels of transformation Purification of genomic DNA. For DNA preparation, the type strains of B. among these species are low (14, 16, 21). However, the se- popilliae and B. lentimorbus were cultured in J medium by using the protocol of quence differences among these species are accompanied by Gordon et al. (6); all other strains were cultured in brain heart infusion medium very few phenotypic differences. B. subtilis can be distinguished (Difco). Genomic DNA for use in transformation experiments was obtained from a spontaneous Rif mutant of each donor strain, as described previously (3). phenotypically from B. mojavensis only by its fatty acid com- The procedure used to prepare highly purified DNA for reassociation experi- position (16) and from B. atrophaeus only by its pigmentation ments has been described previously (13). (12). B. subtilis can be distinguished from B. amyloliquefaciens Transformation. Using the protocol of Cohan et al. (3), we induced strains to by only three phenotypic traits (9, 11) and from the more become competent for transformation in liquid cultures. Each strain was then transformed to rifampin resistance with genomic DNA from its own Rif mutant distantly related organism B. lichenifomis by only five pheno- (homogamic transformation) and with DNAs from Rif mutants of other strains typic traits (9). The dearth of diagnostic phenotypic character- (heterogamic transformation). We calculated transformation frequencies by de- istics in this group suggests there may be species which are termining the fraction of cells from each recipient culture that were transformed closely related to B. subtilis that have not been discovered yet to rifampin resistance, after we corrected for mutation. Sexual isolation (i.e., resistance to transformation) between a recipient and a test donor was quantified (16). Indeed, two close relatives of B. subtilis (B. atrophaeus by determining the ratio of the frequency of homogamic transformation to the and B. mojavensis) have been discovered in only the last 6 years frequency of heterogamic transformation (14, 21). (12, 16). DNA-DNA reassociation and G+C content determinations.The renaturation A recent survey of genetic variation among soil isolates rates of genomic fragments from pairs of strains were determined spectropho- tometrically with a model 250 UV spectrophotometer (Gilford Systems, CIBA- having B. subtilis-like phenotypes revealed an unexpected clus- Corning Diagnostics Corp., Oberlin, Ohio) equipped with a model 2527 ther- ter of strains very closely related to B. subtilis (15). On the basis moprogrammer (13). The equation of De Ley et al. (4) was used to calculate the of the results of a restriction digestion analysis of the DrA, extent of DNA-DNA reassociation. polC, and rpoB genes, a cluster of five strains obtained from The G+C content was estimated by the thermal melting procedure described by Mandel and Marmur (10). Escherichia coli DNA with a G+C content of 51 Death Valley formed a monophyletic group that was most mol% was used for comparison. closely related to B. subtilis (Fig. 1). In this paper we present Phenotypic characterization. Physiological, morphological, and metabolic evidence that these Death Valley group strains are members of characteristics (except fatty acid composition) were determined as described a previously unidentified Bacillus species. previously (6,13). The whole-cell fatty acid composition was determined by using the MIDI system of Sasser (18) and organisms that had been grown for 24 h at 28°C on Trypticase soy agar. We determined the fatty acid compositions of the MATERIALS AND METHODS five members of the Death Valley group and compared these compositions with data for related species obtained previously (16). B. lentimorbus and B. popilliae Bacterial strains. The five strains previously identified as members of the were not characterized because their phenotypes were known to be very different Death Valley group (15) are listed in Table 1. These strains were isolated from from the phenotypes of B. subtilis-like organisms (2). soil obtained at four collection sites in Death Valley (15). A prototrophic, rifampin-susceptible derivative of B. sub& 168, designated strain 1A2 (= NRRL B-14819), was obtained from the Bacillus Genetic Stock Center. The RESULTS Agricultural Research Service Culture Collection at the National Center for Sexual isolation between the Death Valley group and B. Agricultural Utilization Research provided the following type strains: B. amyloliquefaciens NRRL B-14393 (= ATCC 23350), B. atrophaeus NRRL NRS- subtilis. Each Death Valley group strain failed to show any 213, Bacillus lentimorbus NRRL B-2522 (= ATCC 14707), B. lichenifomis evidence of transformation toward rifampin resistance, NRRL NRS-1264 (= ATCC 14580), B. mojavensis NRRL B-14698, Bacillus whether the donor DNA was from the recipient's own Rif popilliae NRRL B-2309 (= ATCC 14706), and B. subtilis NRRL NRS-744 (= mutant or from a Rif mutant of B. subtilis 1A2; that is, the ATCC 6051). (The prefix NRRL indicates that a strain is maintained in the Agricultural Research Service Culture Collection; each NRRL designation in- frequency of appearance of rifampin-resistant colonies when donor DNA was added was about the same as the frequency of appearance under mutation alone (data not shown). Because * Corresponding author. Phone: (517) 432-3771. Fax: (517) 432- of this apparent lack of competence in the Death Valley group 3770. Electronic mail address: [email protected]. strains, sexual isolation between the Death Valley group and 470 VOL.46, 1996 BACILLUS VALLISMORTIS SP. NOV. 471 TABLE 2. Levels of sexual isolation when B. subtifis 1A2 was the recipient and Death Valley group strains were the donors Sexual Donor strain Transformation frequency" isolation B. subtilis 1.42 valueh ~ ~~~~ B. subtilis 2A2 Death Valley group strains DV1-A- 1 -2.39 t 0.12 0.22 ? 0.13 B. mujuvensis RO-H-I~ DVl-F-3= -2.54 t 0.10 0.36 ? 0.15 L DV4-D-3 -2.36 '-+ 0.09 0.18 t 0.16 DV7-C-1 -2.34 -+ 0.10 0.16 t 0.17 DV8-1.7-4 -2.24 2 0.06 0.06 t 0.15 Meanc -2.37 k 0.05 0.20 2 0.0Y' Transformation frequencies were log,, transformed as described by Cohan et al. (3). The values are the mean 2 standard error for each strain, based on five experimental trials. Factor by which the recipient's transformation frequency was reduced in heterogamic transformation compared with the frequency at which the recipient was transformed by its own Rif' mutant's DNA. Sexual isolation values were log,, transformed. The values are the mean 2 standard error for each strain, based on five experimental trials. The recipient was transformed by its own Rif mutant's DNA at an average loglo-transformed frequency of -2.18 ? 0.14 (= 6.64 X lop3) in five experimental trials. ' Mean ? standard error for five Death Valley group strains. P < 0.01 as determined by a one-tailed t test of the difference between the experimental value and zero. other taxa could be studied only when Death Valley group strains were used as donors. We investigated sexual isolation between the Death Valley variance (F4,24< 1). The extents of sexual isolation between group and B. subtilis by using B. subtilis 1A2 as the recipient. In the Death Valley group strains and strain 1A2, although mod- homogamic transformation experiments, B. subtilis 1A2 was est, were highly significant (t = 4.09, P < 0.01; one-tailed t test transformed at a frequency of 6.64 X The frequency of for the difference between the mean log,,-transformed sexual transformation was lower than this value when any of the five isolation value, 0.20 2 0.05, and zero). In contrast, previous Death Valley group strains was used as the donor; strain 1A2 results have shown that the recipient which we used, B. subtilis was transformed by Death Valley group strains at an average 1A2, is not sexually isolated from B.